2. Measurement Uncertainty
LSL USL
Measurement
point pass or
reject?
• Imagine you measure a component and find it to be at the point
shown within the specification limits
• Would you pass or reject the component?
3. Measurement Uncertainty
LSL USL
Measurement
point pass or
reject?
• You now discover that is has a measurement error associated with it
as shown by the error bar above
• Would you pass or reject the component?
4. Main sources of variation
• Materials
M th d• Methods
• Machines
• People
• Environment
5. Measurement System Analysis - MSA
• A scientific and objective method of analysing the validity
of a measurement systemof a measurement system.
• A tool which quantifies
– Equipment variationq p
– Appraiser (operator) variation
– The total variation of a measurement system
E amples of meas rement s stems• Examples of measurement systems
– Micrometer
– Shadowgraphg p
– Go/no-go gage
– Data collection form
S r e– Survey
– On-time delivery report
8. 1. Resolution
D fi iti Th bilit t
What is the length of this
component to the nearest 10th?
• Definition: The capability to
detect the smallest acceptable
change
C I d t
1 2 3 4
Poor resolution• Cause: Inadequate
measurement units
• Guideline “10 bucket rule”
Poor resolution
1 2 3 4
– Increments in the measurement
system should be one tenth of the
product specification or process
variation
Better resolution
What percent of requests are issuedvariation
• Actions:
– Change measurement device
– Record sample averages
Customer Name ________________
Date Received ________________
Date Issued ________________
within 4 hours of receipt??
Record sample averages
– Live with it but understand its
limitations Customer Name ________________
Date Received ________________
Time Received ________________
Poor resolution
R l ti 1/10th Date Issued ________________
Time Issued ________________
Better resolution
Resolution 1/10th
9. 2. Accuracy /Bias
A /Bi Diff
Master value
Reference standard
• Accuracy/Bias: Difference
between the observed average
value and the master reference
XX
X X
• Master value is an accepted,
traceable reference standard
• Actions:
X
X
XX
X
X
X
X
Less accurate• Actions:
– Calibrate regularly
– Use operations instructions
R i ifi ti f
Less accurate
Master value
Reference standard
– Review specifications for
resolution
– Validate data systems input
accuracy
X
XX
X
X
X
accuracy
– Create operational definitions
XX
X
X
X
More accurateMore accurate
10. 3. Linearity
M i “ ” d/• Measurement is “true” and/or
consistent across the range of
the “gauge”
uredegC
• Actions
– Check gauge specification
Rebuild/replace gauge
Linear
range
Temperat
– Rebuild/replace gauge
– Use within restricted range
– Use correction factor
Gauge reading deg C
Form Linearity
Super Outstanding 10Super Outstanding 10
Outstanding 9
Incredible 8
Excellent 7
Great 6Great 6
Very Good 5
Good 4
OK 3
Fair 2Fair 2
Poor 1
11. 4. Stability
M i
Time 1 Time 2
• Measurements remain
constant and predictable over
time i.e. accuracy remains
Shift
constant
• No drifting, sudden shifts or
cyclescycles
• Actions
– maintain and service
equipment
– use control charts
– use SOP
– ensure adequate training
– regular audits
12. 5. Precision – Repeatability & Reproducibility
R t bilit i ti th t
Master value
Reference standard
• Repeatability - variation that
occurs when repeated
measurements are made of the
same item nder identical X
XX
X
X
X
same item under identical
conditions
• Actions
Poor precision
X
XX
X
X
X
– repair, replace, adjust
– SOP
• Reproducibility – variation that
Poor precision
Master value
Reference standard
results when different conditions
are used to make the same
measurements
XXXX
XXX
XX
X• Actions
– training
– SOP
Good precision
XXXXX
Good precision
13. Gauge R&R Studies
• Method of assessing Repeatability & Reproducibility of a
measurement systemmeasurement system
• A number of appraisers (usually two or three) measure a
number of parts (or process output) (usually 5 to 20) anumber of parts (or process output) (usually 5 to 20) a
number of times (usually two or three)
• The results are compared within each appraiserThe results are compared within each appraiser
(Repeatability) and between appraisers (Reproducibility)
• Randomisation is critical for repeat measurements top
avoid learning or copying.
14. Gage R&R Study – Continuous Data
Three appraisers, two
measurements eachmeasurements each
% of Tolerance
% of Total Variation
EV=Equip’t Variation
AV=Appraiser Variation
PV=Part Variation
R&R = √(EV²+AV²)
15. Gage R&R Study – Attribute Data
Two appraisers, two
hmeasurements each
Appraiser Variation
Appraiser vs Control
AV=Appraiser Variation
AC=Appraiser vs Control
R&R = √(AV*AC)
16. Measurement Error Matching Exercise
X X
Time 1
Time 2
A. B.
1. Resolution/Discrimination
2. Accuracy (bias)
X
X
XX
X
X
X
X
X
Shift
A. B.
y ( )
3. Linearity
X X
4. Stability (consistency)
5. Precision – Repeatability &
Reproducibility
degC
EDC
mperatured
XXXX
XXX
XX
X
E.D.C.
1 2 3 4
Gauge reading deg C
Tem
XXXX